NVIDIA GEM-X SOMA regression denoiser needs about 1.1 GB of memory at F32 (a 0.4 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 0.2 GB (F32) to 3.4 GB (F32).
NVIDIA GEM-X SOMA regression denoiser for 3D human motion
NVIDIA GEM-X SOMA is a transformer model that reconstructs full‑body 3‑D motion from monocular video. It outputs 77‑joint poses and global motion trajectories in SOMA format. The model is ready for commercial use under the NVIDIA Open Model License Agreement.
Good at
- Motion capture from video
- Character animation from real footage
- Digital human avatar control
Limits
- Context length 120 tokens
- Requires GPU or CPU with GGML runtime
- No LLM language support
Best for
- Motion capture for games
- Animation for film
- Biomechanical analysis in sports
Who it’s for. Professionals and researchers in machine learning, computer vision, gaming, animation, and biomechanics. The model supports motion capture, character animation, and digital human applications.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| F32 | 0.2 GB | 0.7 GB | Fits | Get file → | |
| F32 | 0.4 GB | 0.9 GB | Fits | Get file → | |
| F32 | 3.4 GB | 3.9 GB | Fits | Get file → |
For 8 GB VRAM, use the 0.18 GB F32 file (est. 0.9 GB memory at 8k context). For 4 GB VRAM, the 0.4 GB F32 file (est. 1.1 GB memory) is suitable. Larger models exceed typical consumer VRAM limits.
What GPU do you need for NVIDIA GEM-X SOMA regression denoiser?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is F32 (0.4 GB file, about 1.1 GB in use).
- 6 GB VRAM: the best quant that fits is F32 (3.4 GB file, about 4.1 GB in use).
Which NVIDIA GEM-X SOMA regression denoiser quant should you choose?
F32 (0.4 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
If memory is tight, the smallest file is F32 at 0.2 GB, with the quality trade-offs that come with fewer bits. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about NVIDIA GEM-X SOMA regression denoiser
How much VRAM does NVIDIA GEM-X SOMA regression denoiser need?
About 1.1 GB at F32 with an 8K context window. That is the 0.4 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run NVIDIA GEM-X SOMA regression denoiser on an 8 GB GPU?
Yes, with F32: the 3.4 GB file needs about 4.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of NVIDIA GEM-X SOMA regression denoiser GGUF?
It depends on the quant: F32 is 0.2 GB, F32 is 0.4 GB, F32 is 3.4 GB.
What context length does NVIDIA GEM-X SOMA regression denoiser support?
The GGUF header lists a trained context length of 120 tokens (120). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.0 GB at 120.
Where can I download NVIDIA GEM-X SOMA regression denoiser GGUF?
The files are on Hugging Face in the LocalAI-io/GEM-X-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does NVIDIA GEM-X SOMA regression denoiser use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run this model on 8 GB VRAM?
Yes, the 0.18 GB F32 file uses about 0.9 GB memory at 8k context, fitting within 8 GB VRAM.
What context length does it support?
The model was trained with a context of 120 tokens.
Is it licensed for commercial use?
Yes, it is licensed under the NVIDIA Open Model License Agreement for commercial use.